Agent Vi launches enterprise solution

TEL AVIV, Israel—Agent Video Intelligence, a global provider of video analytics solutions, recently announced the launch of innoVi Enterprise, a deep learning-powered video analytics software designed to meet the high-end security and safety needs of multiple vertical sectors.

“The importance of innoVi Enterprise being centrally managed—whether Agent Vi hosted or customer hosted—is paramount,” AJ Frazer, VP business development, cloud services at Agent Vi, told Security Systems News. “innoVi's modern software approach was designed with IT environments in mind, using the latest high availability, scalable micro-services architecture. This ensures the software is seamlessly and constantly updated for new functionality, cybersecurity updates, bug fixes, and the like, with no system stoppage.”

Any ONVIF/RTSP cameras can connect to innoVi via the innoVi Edge appliance, which transforms any camera into an intelligent, video analytics-enabled device. “Enterprise users can easily connect cameras to the innoVi service, and scale from small to large seamlessly without having to redesign their systems,” Frazer noted.

Frazer pointed out that central management also takes into consideration the internal operational factors faced by most organizations. “Overall system costs are reduced because we, Agent Vi, bear the cost of hardware, space, processing power, UPS, cooling, routine server maintenance, computing and storage,” he explained. “Customers can use their operating budgets (OPEX) instead of capital budgets (CAPEX), which is a huge benefit.”

The innoVi Enterprise solution includes a comprehensive suite of real-time detection rules as well as an autonomous real-time Anomaly Detection capability. Its Anomaly Detection capability, which does not require any pre-definition or configuration by the operator, is a continuously-learning capability that instantly alerts users to irregular incidents, particularly in complex scenarios or very large installations where implementation of standard detection rules is not applicable.

"This next-generation software approach also enables us to offer a whole new level of analytics both in terms of performance and new features like our self-learning Anomaly Detection capability,” Frazer said. “Anomaly Detection's deep learning-driven algorithms can quickly learn scenarios and alert to unusual events, without any rule definition or configuration. That's particularly valuable for large-scale deployments where the sheer volume of cameras prevents the implementation of standard detection rules.”

Powered by Deep Learning technology, Frazer said that innoVi Enterprise “boasts a highly accurate” object classification methodology that learns to accurately classify and distinguish between various object types such as persons, cars, motorcycles, bicycles, trucks, buses and static objects. This allows users to fine-tune system alerts to very specific objects of interest and opens the door to much broader opportunities for implementation of video analytics, especially in large-scale outdoor surveillance installations.

innoVi Enterprise eliminates the need for continuous human monitoring of surveillance cameras. “It automatically detects and alerts to security and safety incidents in real-time, helping users to uncover otherwise hidden events, generate insights, and provide situational awareness to effectively manage incidents and enable effective responses, as events unfold,” the company said.